My 1934 GE "CA" Monitor Top.

Thanks for the link. As Chris says, the heater needs to be checked. Also, the way that rattle sounds makes me think of NCG's as well as the heater.
My CA-2 never rattles except very quietly, for a few seconds, only when the room temp gets below about 15C, and then only rarely.
However, my CA-1 does rattle if the room temp is below about 18C. I'm not sure why the difference between the two, but one thing I note is the original oil heater (in the CA-1) is shorter than my homemade one for the CA-2. Anyway, rattling in a cold room is a known characteristic and mentioned in the manual, so I'm not worried about it. It should never rattle in a warm room.
Your off time is too short - that needs to be looked into, but first check the heater and NCG's.
Regarding the supply voltage, the official is 115V 60c/s (although the label says 110V). Although 130V is within GE's spec, it's obviously going to run cooler at 115V, which is a good thing. So, your variac idea is worthwhile. Reducing the voltage slightly won't reduce the speed - it's a synchronous induction motor and speed depends on mains frequency.

Jerry,

Your're in luck. Cablehack is the king of NCG (non-condensable gas).

Not sure if that's a compliment or an insult.

Put it this way, the Bristol key is always in easy reach, and the bottle of oil is next to the CA-2. Seriously though, the CA-2 purging is thankfully much, much less than it used to be.

Jerry,
NCG's are non condensable gases. A characteristic of the CA units is that part of them always works in a vacuum. So, if there's a leak, air will enter the unit, rather than refrigerant leak out. Air is one form of NCG. The other comes about from the breakdown of the methyl formate - something unique to CA's. The methyl formate breaks down if there's any moisture in the system (like my damaged unit being left open). The by products of the breakdown are various gases which won't condense under pressure. It seems that methyl formate breaks down anyway, even if not deliberately contaminated. The end result is after a period of time, which can be many years, the non condensables have to be purged. They cause problems by forcing the float valve shut - the result being a lower than normal frost line, and a higher compressor pressure (hence the rattle). They are also undesirable because they reduce heat radiating area from the condenser.
The reason for the CA only being extant for two years, despite the safer refrigerant, is because of the headache methyl formate caused GE once it started to break down.

I'd suggest having a read through the manual regarding the purging procedure members.iinet.net.au/~cool386/ge_service/CA_factory_manual.zip, and also the CA notes in this book members.iinet.net.au/~cool386/ge_service/sealed_units.zip. Here there is a pictorial explanation of NCG's, and how to purge them from the CA unit. I think it's particularly well written and the notes cover most CA problems in simple terminology.

One problem the manuals don't mention is what happens when the needle and seat inside the float chamber wear. This needle and seat wear is another unique to CA characteristic, because these parts have been found to have been replaced in rebuilt units opened up by other forum members. Original examples often have strange notches and patterns worn into them. The result is when these parts wear, the evaporator cannot get as cold as it should (insufficient pressure differential between the hot liquid refrigerant and the evaporator), and cycling times become poor. Fortunately, by installing a capillary tube in its place, the problem is eliminated permanently, and performance is restored to as good as new.

Jerry, I'd bleed the NCG'S as outlined in the manual. It's a very simple procedure that requires only a $5 bristol key. It made a world of difference on my CA form A. I'd hate to see you get rid of that top. When I had it apart for the rewire, I found it to be in exceptional condition. If you like, give me a call. I'll gladly walk you through it. Allan and Cablehack explained the process to me And there's nothing to it.

I don't really understand why you're so mad at an old refrigerator. Just chill out and go to bed and deal with it tomorrow.

Something's not right - it just requires a certain approach. Even when I have deliberately filled my CA with NCG in the form of nitrogen it is all out in much, much less than 4 hours. Longest I have ever purged is about 1.5hrs. I'm not saying it's happened in this case as I don't know how the purging has been done, but after 4hrs I'd be worrying that refrigerant has been lost.
The long run time makes me wonder about this because with loss of refrigerant, the evaporator will not frost where the thermostat's capillary tube is clamped to it. It will take forever to get cold enough to switch off the compressor.
By no means is the unit 'junk'. The compressor runs, it appears to have a useable float valve - and even it if didn't, that's easily dealt with. Another thing in its favour is it hasn't been open to contamination. If it wasn't for the distance, I'd gladly have it.